Calcium Channel Blockers and Risk of Breast Cancer: A Meta-Analysis of 17 Observational Studies Wen Li1", Qi Shi1, Weibing Wang2, Jianrong Liu1, Qi Li3, Fenggang Hou1* 1 Oncology Department of Shanghai Municipal Hospital of Traditional Chinese Medicine affiliated to Shanghai TCM University, Shanghai, China, 2 Department of Epidemiology, School of Public Health, Fudan University, Shanghai, China, 3 Oncology Department of Shuguang Hospital, Shanghai University of Traditional Chinese Medicine, Shanghai, China
Abstract Purpose: Studies on the association between the use of calcium channel blockers (CCBs) and breast cancer risk have reported inconsistent results. We quantitatively assessed this association by conducting a meta-analysis based on the evidence from observational studies. Methods: We searched PubMed, MEDLINE, EMBASE and the Cochrane Library for relevant studies published up to and including December 31, 2013. We calculated pooled risk ratios (RRs) for cancer risk. Results: A total of 17 studies (9 cohort studies, 8 case-control studies) were selected for further study. These studies included 149,607 female subjects, of which 53,812 were CCBs users, who were followed for 2–16 years. The risks of breast cancer among patients receiving CCBs were significantly different for the pooled RRs (95% confidence interval) of cohort studies 1.08 (0.95, 1.20) and case-control studies 0.98 (0.86, 1.09). Differences were also noted for cancer risk, for CCBs use of ,5 years 0.96 (0.78, 1.15), and for .5 years 1.01 (0.74, 1.28), as well as for ever used 1.08 (0.95, 1.20), and for current use 1.13 (0.83, 1.42). The RR for studies longer than 10 years was 1.71 (1.01, 2.42), and for studies evaluating nifedipine was 1.10 (0.87, 1.33) and diltiazem was 0.75 (0.40, 1.10). Conclusions: The long-term use of CCBs appears to have a significant relationship with breast cancer. Well-designed clinical trials are needed to optimize the doses and types of these drugs needed to minimize their carcinogenic potential. Citation: Li W, Shi Q, Wang W, Liu J, Li Q, et al. (2014) Calcium Channel Blockers and Risk of Breast Cancer: A Meta-Analysis of 17 Observational Studies. PLoS ONE 9(9): e105801. doi:10.1371/journal.pone.0105801 Editor: Rodney John Scott, University of Newcastle, Australia Received March 4, 2014; Accepted July 25, 2014; Published September 3, 2014 Copyright: ß 2014 Li et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work is supported by Innovation Project of Shanghai Municipal Education Commission (12YZ057), the Fundamental Research Funds for the Central Universities (78210021) and the Chenguang Program from Shanghai Municipal Education Commission (10CG25). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * Email: [email protected] " WL is the first author on this work.
CCBs, which can inhibit apoptosis and thus facilitate the division of cells with malignant potential [12], have been found to increase breast cancer risk in some studies. In 1996, Pahor et al., found that in patients who had received CCBs the risk of breast cancer was increased [13]. However, the data is equivocal with some studies lending support to this association between CCBs and breast cancer risk [14–17], but not others [18–23]. Therefore the debate has been fuelled by conflicting data. We conducted a meta-analysis of observational studies of CCBs, to examine their effect on the occurrence of breast cancers.
Introduction Breast cancer is by far the most common cancer diagnosed in women (ranking second for both sexes combined) and the most common cause of death in women (ranking fifth for both sexes combined) worldwide [1,2]. The age standardized incidence increased by 40% from 1979 to 1992 in England and Wales [3]. Breast cancer is now also the leading cause of cancer-related mortality among females in economically developing countries, a shift from the previous decade during which the most common cause of cancer death was cervical cancer [4]. Hence, there is a growing interest in breast cancer prevention and several strategies have been evaluated with some promising results. Calcium channel blockers (CCBs) are prescribed primarily for treatment of hypertension and coronary heart disease [5,6]. CCBs are also used to treat esophageal diseases [7]. These drugs are potent drugs that affect various organ systems, and can cause constipation [8], increase the risk of hemorrhage [9], and impair differentiation during embryogenesis [10]. Furthermore, cases of lupus after use of diltiazem have been reported [11]. PLOS ONE | www.plosone.org
Methods Search This meta-analysis was conducted in accordance with PRISMA(Preferred Reporting Items for Systematic Reviews and MetaAnalyses) guidelines [24,25]. A comprehensive literature search was carried out using PubMed database, MEDLINE, EMBASE, China Academic Journals Full-text Database, and the Cochrane Library. We used the term ‘‘breast cancer’’ or ‘‘cancer’’ or 1
September 2014 | Volume 9 | Issue 9 | e105801
A Meta-Analysis of 17 Observational Studies
‘‘cancer(s)’’ or ‘‘neoplasm(s)’’ or ‘‘malignancy(ies)’’ or ‘‘carcinoma’’ in combination with ‘‘calcium channel blockers’’ or ‘‘verapamil’’ or ‘‘diltiazem’’ or ‘‘nifedipine’’ or ‘‘dihydropyridines’’ or ‘‘amlodipine’’ to identify the studies related to CCBs and breast cancer risk. The reference lists of all eligible articles and reviews were also scanned to identify additional relevant studies. All cancer studies were included to ensure that no results for breast cancer were overlooked that might have been reported as part of a larger study that included other cancer types.
performed according to (i) study design (cohort and case-control), (ii) duration of use, (iii) time to study observation (ever and current), and (iv) drug type to examine the impact of these factors on the association between CCB use and breast cancer. To test the robustness of this association, we performed a sensitivity analysis by excluding one study at a time. Cumulative meta-analysis was also performed to identify the change in trend of reporting risk over time.
Results
Selection/Study Characteristics
Search results
The studies considered in this meta-analysis were all observational (cohort or case-control) studies that evaluated exposure to CCBs and risk of breast cancer. We included all articles irrespective of publication length; for example, articles published as short reports or conference abstracts, even though the critical appraisal of such publications is limited, were included. Articles were excluded from the analysis if they had insufficient published data for determining an estimate of relative risk (RR) and confidence interval (CI). When there were multiple publications from the same population, only data from the most-recent report were included in the meta-analysis and remaining data were excluded. Studies reporting different measures of RR [e.g. risk ratio, rate ratio, and odds ratio (OR)] were included in the metaanalysis. In practice, these measures of effect yield a similar estimate of RR, since the absolute risk of breast cancer is low.
A total of 128 abstracts were identified and screened, and 108 studies were reviewed in detail. After excluding laboratory studies, ecological studies, and reviews, 20 studies were initially selected for analysis [13–23,29–37]. Three of these studies were excluded because they did not provide sufficient information [35–37], leaving 17 in the final analysis (Figure 1) [13–23,29–34].
Qualitative Summary The characteristics of the selected studies are shown in Table 1. Overall, 17 relevant studies (9 cohort and 8 case-control) published between 1996 and 2013 were identified. A total of 149,607 female subjects, including 53,812 CCBs users, were enrolled in these studies and followed for 2 to 16 years. Nine studies were from the United States [13,14,16,17,19–21,31,33], four from the UK [15,18,22,29], two from Denmark [23,30], and two from Canada [32,34]. Nine studies were cohort studies [13,15,18– 20,23,30,33,34], while seven were case-control studies [14,17,19,21,22,29,31], and one was a nested case-control study [32]. Of the 17 studies [13–23,29–34], 7 were population-based and 10 were hospital-based. All studies were controlled for potential confounding factors (at least for age) by matching or adjustments. Further, 7 studies [14,17–19,29,31,32] reported OR estimates for the association between CCB use and the risk of breast cancer, while 10 [13,15–17,20–23,30,34] studies reported RR. All studies had a median follow-up time of at least 3 years, while 2 studies had a median follow-up of more than 10 years.
Data abstraction We reviewed each full-text report to determine its eligibility and extracted and tabulated all the relevant data independently. The extracted data included the characteristics of the subjects (including age, number and treatment), study design, published year, follow-up period, journal of article and the covariate adjustment. Study authors were contacted as needed to obtain detailed data. Any disagreement was resolved by a consensus among the investigators.
Quantitative data synthesis The RRs and ORs with their 95% CIs were extracted from individual studies for the highest versus the lowest quartiles, and the OR from case-control studies were assumed as the estimate of the RR value in the CCBs intake meta-analysis studies. Both the fixed-effects and random-effects models were used to calculate the pooled estimate and its 95% CI. If more than one risk estimate was provided in a study that had been stratified by covariates, the estimates were pooled before data were entered into the final analysis. The heterogeneity of the data was quantified by the Q statistic and in combination with the I2 statistic, which represents the percentage of variability across studies that is attributable to heterogeneity rather than chance. Heterogeneity among studies was considered significant when P,0.05 for the Q statistic or when the I2 value was more than 50%. If there was significant heterogeneity among the studies, the random-effects model was used, otherwise, the fixed-effects model was acceptable. Publication bias was represented by funnel plots and was further assessed by the Egger test and Begg test [26]. When there was a significant publication with regard to CCB intake and breast cancer risk, the trim and fill method was applied to correct the publication bias [27,28]. All the statistical analyses were performed with STATA12.0 (StataCorp, College Station, TX, USA). We first estimated the risk of breast cancer in CCB users compared to the non-users. We used the available data from previously collected studies, which reported RR estimates for these particular associations. Pre-specified subgroup analyses were PLOS ONE | www.plosone.org
Figure 1. Flow diagram of studies included in the metaanalysis. doi:10.1371/journal.pone.0105801.g001
2
September 2014 | Volume 9 | Issue 9 | e105801
PLOS ONE | www.plosone.org
case control
cohort
cohort
Rosenberg et al. [21] USA, 1998
Michels et al. [20] US, 1998
Sørensen et al. [19] USA, 2000
3
September 2014 | Volume 9 | Issue 9 | e105801
case control
nested case-control
cohort
case control
cohort
Davis et al. [29] USA, 2007
Assimes et al. [30] Canada, 2008
Saltzman et al. [31] USA, 2013
Li et al. [13] USA, 2013
Holmes et al. [32] Canada, 2013
Cancer Epidemiology
JAMA Internal Medicine
5
.10
12
7
pharmacoepidemiology and drug safety Cancer Causes Control
3
5.7
7
3
5
3.2
6
16
5
3
5
5
Follow-up (years)
Eur J Epidemiol
BreastCancer Research and Treatment
pharmacoepidemiology and drug safety
CANCER
Arch InternMed
CANCER
CANCER
JAMA
J Hypertens
CANCER
Hypertension
THE LANCET
THE LANCET
Journal of article
63
55–74
$65
71.8
20–74
50–67
30–79
65–79
.59
66.4
60.8
40–69
51.7
72.4
NA
71.3
79.3
Age (years)
4019
2851
3201
77887
1247
49950
21177
1001
925
11726
18635
6641
2770
3198
17911
1096
3259
Number of patients
4019
1960
392
1136
600
4381
3333
512
190
11726
2361
2893
1129
759
17911
221
289
No.cases
NA
Dihydropyridines Diltiazem Verapamil Nifedipine
NA
NA
NA
Dihydropyridines Diltiazem Verapamil Nifedipine
Amlodipine Nifedipine Diltiazem
Diltiazem Verapamil Dihydropyridines
Dihydropyridines Diltiazem Verapamil Nifedipine
Dihydropyridines Diltiazem Verapamil
NA
Verapamil, Nifedipine Diltiazem
Dihydropyridines Diltiazem Verapamil
Verapamil, Nifedipine Diltiazem
Dihydropyridines Diltiazem Verapamil
Amlodipine Verapamil, Nifedipine Diltiazem
Verapamil Diltiazem Nifedipine
Study drug
AHT, antihypertensivemedications, HRT, hormone replacement therapy, BMI, body mass index, NA, not available, NSAID,nonsteroidal antiinflammatory drugs. doi:10.1371/journal.pone.0105801.t001
cohort
Fryzek et al. [28] Denmark,2006
case-control
cohort
Hole et al. [15] UK, 1998
Gonza´lez-Pe´rez et al. [27] UK, 2004
cohort
Fitzpatrick et al. [16] USA, 1997
case-control
cohort
Olsen et al. [23] Danish, 1997
Li et al. [14] USA, 2003
case control
Jick et al. [22] UK, 1997
case control
cohort
Pahor et al. [13] USA, 1996
Meier et al. [18] UK, 2000
Analysis design
References Country, year
Table 1. Characteristic of studies included in the meta-analysis.
age, stage at diagnosis history of previous cancer, and urban/rural residence
age, reference year, county, race/ ethnicity, and recency of alcohol use
age, alcohol use, income, age at menopause, waist-hip ratio; all compared against common reference group of never AHT users
sex, year of entry, age at entry, and type of antagonist
sex, year of entry, age at entry, and type of antagonist
age, calendar period, HRT use, NSAID use, parity, and age at first birth
age, calendar year, BMI, alcohol intake, smoking status, HRT use, prior breast lump and/or biopsy, hypertension
age,BMI, receipt of HRT, alcohol use, smoking status
age BMI,smoking status, alcoholism,hysterectomy, breast lumps
sex, year of entry, age at entry, and type of antagonist
history of benign breast disease, age at menarche, parity, age at first birth, and age at menopause
sex, year of entry, age at entry, and type of antagonist
sex, year of entry, age at entry, and type of antagonist
age, race, parity, age at menopause,and self-reported diabetes
sex, year of entry, age at entry, and type of antagonist
age, dose,and duration of use
age, sex, ethnic origin,heart failure, number of hospital admissions, cigarette smoking, and alcohol intake.
Covariate adjustment
A Meta-Analysis of 17 Observational Studies
A Meta-Analysis of 17 Observational Studies
a statistically contradictory effect revealed (summary RR = 1.71, 95%CI: 1.01, 2.42, for Pheterogeneity = 0.379, I2 = 0.0%). Furthermore, when subgroup analysis was performed according to different drugs, nifedipine and diltiazem the pooled RR for the studies which used nifedipine was 1.10 (95% CI: 0.87, 1.33) and studies which used diltiazem was 0.75 (95%CI: 0.40, 1.10). As shown in Figure 7, there was no evidence of heterogeneity (Q = 9.47, P = 1.000, I2 = 0.0% and Q = 4.23, P = 0.627, I2 = 0.0%, for nifedipine and diltiazem, respectively).
Quantitative Summary (Meta-analysis) We found no association between CCB use and risk of breast cancer in 12 studies (RR = 1.02, 95%CI: 0.94, 1.11) presented in Figure 2. However, there was high evidence of a lack of heterogeneity among these studies (Pheterogeneity = 0.168, I2 = 28.2%). Stratification by study design showed that the association was neutral in cohort studies (RR = 1.08, 95%CI: 0.95, 1.20) and there was a non-significant inverse association in case-control studies (RR = 0.98, 95% CI: 0.86, 1.09) [P = 0.02]. Moreover, in the subgroup analysis based on the duration of drug use, no differences were detected in duration of CCB use and breast cancer risk: ,5 years (RR = 0.96, 95%CI: 0.78, 1.15), .5 years (RR = 1.01, 95%CI: 0.74, 1.28), and ever used (RR = 1.08, 95%CI: 0.95, 1.20). However, the pooled RR for current use was 1.13 (95%CI: 0.83, 1.42) (Figure 3). The combined RR was 1.00(95%CI: 0.99,1.00) for the 3 studies which followed patients for .10 years indicating that there was no association between CCB use and breast cancer (I2 = 71.3%, Figure 4). The heterogeneity appeared to be driven by the effect of the study by Hole et al. [15] (Figure 5). When it was excluded from the analysis, the high heterogeneity was eliminated (Figure 6), and
Assessment of Publication bias The funnel plots for the use of CCBs and breast cancer did not reveal evidence of obvious asymmetrical publication bias. Furthermore, the Begg’s funnel plot (Figure 8) based on the selected studies, calculated P = 0.068 in Egger’s test and P = 0.055 egg’s test, respectively. Therefore, there is no evidence of publication bias in the analysis. However, this is indicative of some degree of publication bias, particularly in breast cancer where likely much more data exists on the topic in other studies that did not present these associations in their papers.
Figure 2. The Forest plot of relative ratios (RR) for breast cancer incidence rate between CCB users and non-users in cohort and case-control studies. Horizontal lines represent 95%CI. The diamond (and dash vertical line) represents the overall RR estimate, with the 95%CI given by its width. The solid vertical line is at the null value (RR = 1). doi:10.1371/journal.pone.0105801.g002
PLOS ONE | www.plosone.org
4
September 2014 | Volume 9 | Issue 9 | e105801
A Meta-Analysis of 17 Observational Studies
Figure 3. Meta-analysis and pooled relative ratios (RR) of the time-related CCBs use and breast cancer occurrence. doi:10.1371/journal.pone.0105801.g003
breast cancer. These findings suggest that longer follow-up maybe required to show any association between CCB use and breast cancer. In our subgroup analyses, the results were not substantially affected by study design. Cohort and case-control studies alone showed no association between CCBs use and risk of breast cancer. Nevertheless, there was deviation in the subgroup analyses which evaluated the different CCBs (nifedipine and diltiazem). The test of interaction was not statistically significant in subgroup analyses but was significant among subgroups representing nifedipine use. Although, biological mechanisms through which calcium-channel blockers could influence breast cancer risk are unknown, CCBs may inhibit apoptosis through increasing intracellular calcium. The results that the increased risk with nifedipine, and that an decreased risk associated with diltiazem may help inform studies aimed at elucidating potential biological mechanisms. In short, we found an association between CCB use and breast cancer risk. Moreover, it has been suggested that CCBs increase
Discussion Over the past few decades, there has been increasing understanding of the role of CCBs in the development of breast cancer. The present meta-analysis of 17 observational studies, indicates that there is no increase in breast cancer risk among CCBs users as compared to non-users. This association remained stable even after the sensitivity analysis. Overall, when compared to non-users of CCBs, we found no significant difference in breast cancer risk among ever users, use for ,5 years and use for .5 years. However, an inverse association was observed that suggested there is a 13% increase in the risk of breast cancer occurrence in current CCB users compared to nonusers. There was significant heterogeneity among studies with a duration of longer than 10 years. Nevertheless, the heterogeneity was eliminated when Hole et al. was removed from the analysis [15]. Furthermore, an inverse association was documented and the combined RR was 1.71%. This result provides evidence that longterm use of CCBs may be associated with an increased risk of
PLOS ONE | www.plosone.org
5
September 2014 | Volume 9 | Issue 9 | e105801
A Meta-Analysis of 17 Observational Studies
Figure 4. The combined relative ratios (RR) and 95%CI for the 3 studies which had a duration of longer than 10 years. doi:10.1371/journal.pone.0105801.g004
studies, it is uncertain whether the cases are comparably representative. Moreover, both the funnel plot and Egger’s test do not have high enough power to detect the bias. Also, no Asian was included in our analysis. Finally, the use of CCBs differed across the studies, and some of these studies did not assess or adjust for enough potential confounding variables. However, potential publication bias could be of concern because small studies with null results tended not to be published, especially in the case of clinical trials. In our meta-analysis, we found no evidence of
the risk of cancer by inhibiting apoptosis, or programmed cell death, by DNA fragmentation of dysfunctional and old cells [38]. The CCBs might affect cancer risk generally or be limited to specific sites where calcium mechanisms predominate. Intracellular calcium ion balance regulates apoptosis in vitro [39]; for example, mitochondrial calcium uniporter silencing potentiates caspase-independent cell death in MDA-MB-231 breast cancer cells [40]. In addition, calcium antagonists inhibit apoptosis in vitro [41,42]. Animal data suggest that nifedipine reduced resting calcium concentration and apoptotic gene expression in mice [43]. That is to say, the available laboratory evidence lends support to our findings, which indicate that the use of CCBs increases breast cancer risk. What’s more, it is feasible that breast tissue may be more vulnerable to alterations in apoptotic activity than the other types of tissue. In complex secretory tissue such as the mammary gland, a complex relationship between apoptosis and breast carcinoma exists that may hormonally related. Therefore, more related studies are needed to illustrate this hypothesis.
Strengths and limitations The strength of the present analysis lies in inclusion of 17 observational studies, reporting data from more than 149,607 female participants from multiple nations and was performed with a high level of precision, including 53,812 CCBs user. In addition, this was the first study to use a meta-analysis to investigate the use of CCBs and breast cancer risk. In the analysis of cancer incidence, there was no significant difference in RR between the case-control studies and the cohort studies. Our meta-analysis has several limitations. First, we did not search for unpublished studies or for original data. Second, the included studies were different although we did not detect significant publication bias between PLOS ONE | www.plosone.org
Figure 5. The sensitivity analysis for the 3 studies followed-up for more than ten years. doi:10.1371/journal.pone.0105801.g005
6
September 2014 | Volume 9 | Issue 9 | e105801
A Meta-Analysis of 17 Observational Studies
Figure 6. The combined relative ratios (RR) and 95%CI for the 2 studies followed-up more than ten years. doi:10.1371/journal.pone.0105801.g006
Figure 7. The combined relative ratios (RR) and 95%CI for the 3 studies which state the use of nifedipine or diltiazem. doi:10.1371/journal.pone.0105801.g007
PLOS ONE | www.plosone.org
7
September 2014 | Volume 9 | Issue 9 | e105801
A Meta-Analysis of 17 Observational Studies
Conclusions In conclusion, the long-term use of CCBs appears to have a significant relationship with breast cancer. These findings provide support for the appropriate use of CCBs for those patients who have potentially increased risk of breast cancer. However, more well-designed clinical trials are needed to determine the effect of CCBs on breast cancer, and to optimize the doses and types of these drugs needed to minimize their carcinogenic potential.
Supporting Information Checklist S1 PRISMA Checklist.
(PDF)
Author Contributions
Figure 8. Begg’s funnel plot of studies included in the analysis. The funnel plot displays log odds ratio (OR) against its standard error (s.e.) for each individual study. The horizontal line represents the estimate of the overall OR, with the dash lines indicate the expected 95% CI for a given s.e. doi:10.1371/journal.pone.0105801.g008
Conceived and designed the experiments: WL QS. Performed the experiments: WL JL. Analyzed the data: WL WW. Contributed reagents/materials/analysis tools: WL WW. Wrote the paper: WL QS WW JL QL FH.
publication bias. Therefore, we will update our study when possible.
References 18. Meier CR, Derby LE, Jick SS, Jick H (2000) Angiotensin-converting enzyme inhibitors, calcium channel blockers, and breast cancer. Archives of internal medicine 160: 349. 19. Sørensen HT, Olsen JH, Mellemkjær L, Thulstrup AM, Steffensen FH, et al. (2000) Cancer risk and mortality in users of calcium channel blockers. Cancer 89: 165–170. 20. Michels KB, Rosner BA, Walker AM, Stampfer MJ, Manson JE, et al. (1998) Calcium channel blockers, cancer incidence, and cancer mortality in a cohort of US women. Cancer 83: 2003–2007. 21. Rosenberg L, Rao RS, Palmer JR, Strom BL, Stolley PD, et al. (1998) Calcium channel blockers and the risk of cancer. JAMA: The Journal of the American Medical Association 279: 1000–1004. 22. Jick H, Jick S, Derby LE, Vasilakis C, Myers MW, et al. (1997) Calcium-channel blockers and risk of cancer. The Lancet 349: 525–528. 23. Olsen JH, Sørensen HT, Friis S, McLaughlin JK, Steffensen FH, et al. (1997) Cancer risk in users of calcium channel blockers. Hypertension 29: 1091–1094. 24. Knobloch K, Yoon U, Vogt PM (2011) Preferred reporting items for systematic reviews and meta-analyses (PRISMA) statement and publication bias. Journal of Cranio-Maxillofacial Surgery 39: 91–92. 25. Moher D, Liberati A, Tetzlaff J, Altman DG (2009) Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Annals of internal medicine 151: 264–269. 26. Egger M, Smith GD, Schneider M, Minder C (1997) Bias in meta-analysis detected by a simple, graphical test. Bmj 315: 629–634. 27. Peters JL, Sutton AJ, Jones DR, Abrams KR, Rushton L (2007) Performance of the trim and fill method in the presence of publication bias and between-study heterogeneity. Statistics in medicine 26: 4544–4562. 28. Duval S, Tweedie R (2000) Trim and Fill: A Simple Funnel-Plot–Based Method of Testing and Adjusting for Publication Bias in Meta-Analysis. Biometrics 56: 455–463. 29. Gonzalez-Perez A, Ronquist G, Garcia Rodriguez LA (2004) Breast cancer incidence and use of antihypertensive medication in women. Pharmacoepidemiology and drug safety 13: 581–585. 30. Fryzek JP, Poulsen AH, Lipworth L, Pedersen L, Nørgaard M, et al. (2006) A cohort study of antihypertensive medication use and breast cancer among Danish women. Breast cancer research and treatment 97: 231–236. 31. Davis S, Mirick DK (2007) Medication use and the risk of breast cancer. European journal of epidemiology 22: 319–325. 32. Assimes TL, Elstein E, Langleben A, Suissa S (2008) Long-term use of antihypertensive drugs and risk of cancer. Pharmacoepidemiology and drug safety 17: 1039–1049. 33. Saltzman BS, Weiss NS, Sieh W, Fitzpatrick AL, McTiernan A, et al. (2013) Use of antihypertensive medications and breast cancer risk. Cancer Causes & Control 24: 365–371. 34. Holmes S, Griffith EJ, Musto G, Minuk GY (2013) Antihypertensive medications and survival in patients with cancer: A population-based retrospective cohort study. Cancer epidemiology 37: 881–885. 35. Beiderbeck-Noll A, Sturkenboom M, Van Der Linden P, Herings R, Hofman A, et al. (2003) Verapamil is associated with an increased risk of cancer in the elderly: the Rotterdam study. European Journal of Cancer 39: 98–105.
1. Siegel R, Naishadham D, Jemal A (2012) Cancer statistics, 2012. CA: a cancer journal for clinicians 62: 10–29. 2. Hunter DJ, Reddy KS (2013) Noncommunicable diseases. New England Journal of Medicine 369: 1336–1343. 3. Quinn M, Allen E (1995) Changes in incidence of and mortality from breast cancer in England and Wales since introduction of screening. Bmj 311: 1391– 1395. 4. Jemal A, Bray F, Center MM, Ferlay J, Ward E, et al. (2011) Global cancer statistics. CA: a cancer journal for clinicians 61: 69–90. 5. Chobanian AV (1996) Calcium channel blockers. JAMA: the journal of the American Medical Association 276: 829–830. 6. Psaty BM, Heckbert SR, Koepsell TD, Siscovick DS, Raghunathan TE, et al. (1995) The risk of myocardial infarction associated with antihypertensive drug therapies. JAMA: the journal of the American Medical Association 274: 620– 625. 7. Konrad-Dalhoff I, Baunack A, Ra¨msch K, Ahr G, Kraft H, et al. (1991) Effect of the calcium antagonists nifedipine, nitrendipine, nimodipine and nisoldipine on oesophageal motility in man. European journal of clinical pharmacology 41: 313–316. 8. Pahor M, Guralnik J, Chrischilles EA, Wallace RB (1994) Use of laxative medication in older persons and associations with low serum albumin. Journal of the American Geriatrics Society 42: 50–56. 9. Wagenknecht L, Furberg C, Hammon J, Legault C, Troost B (1995) Surgical bleeding: unexpected effect of a calcium antagonist. BMJ: British Medical Journal 310: 776. 10. Pahor M, Carbonin P, Guralnik J, Havlik R, Furberg C (1996) Risk of gastrointestinal haemorrhage with calcium antagonists in hypertensive persons over 67 years old. The Lancet 347: 1061–1065. 11. Stein G, Srivastava MK, Merker H-J, Neubert D (1990) Effects of calcium channel blockers on the development of early rat postimplantation embryos in culture. Archives of toxicology 64: 623–638. 12. Carson DA, Ribeiro JM (1993) Apoptosis and disease. The Lancet 341: 1251– 1254. 13. Pahor M, Guralnik JM, Ferrucci L, Corti M-C, Salive ME, et al. (1996) Calcium-channel blockade and incidence of cancer in aged populations. The Lancet 348: 493–497. 14. Li CI, Malone KE, Weiss NS, Boudreau DM, Cushing-Haugen KL, et al. (2003) Relation between use of antihypertensive medications and risk of breast carcinoma among women ages 65–79 years. Cancer 98: 1504–1513. 15. Hole DJ, Gillis CR, McCallum IR, McInnes GT, MacKinnon PL, et al. (1998) Cancer risk of hypertensive patients taking calcium antagonists. Journal of hypertension 16: 119–124. 16. Fitzpatrick AL, Daling JR, Furberg CD, Kronmal RA, Weissfeld JL (1997) Use of calcium channel blockers and breast carcinoma risk in postmenopausal women. Cancer 80: 1438–1447. 17. Li CI, Daling JR, Tang M-TC, Haugen KL, Porter PL, et al. (2013) Use of antihypertensive medications and breast cancer risk among women aged 55 to 74 years. JAMA internal medicine 173: 1629–1637.
PLOS ONE | www.plosone.org
8
September 2014 | Volume 9 | Issue 9 | e105801
A Meta-Analysis of 17 Observational Studies
36. Braun S, Boyko V, Behar S, Reicher-Reiss H, Laniado S, et al. (1998) Calcium Channel Blocking Agents and Risk of Cancer in Patients With Coronary Heart Disease 1. Journal of the American College of Cardiology 31: 804–808. 37. Lindberg G, Lindblad U, Lo¨w-Larsen B, Merlo J, Melander A, et al. (2002) Use of calcium channel blockers as antihypertensives in relation to mortality and cancer incidence: a population-based observational study. pharmacoepidemiology and drug safety 11: 493–497. 38. Moon D, McCormack D, McDonald D, McFadden D (2013) Pterostilbene induces mitochondrially derived apoptosis in breast cancer cells in vitro. Journal of Surgical Research 180: 208–215. 39. Handayani S (2013) Selaginella Active Fractions Induce Apoptosis on T47D Breast Cancer Cell. Indonesian Journal of Pharmacy 23. 40. Curry MC, Peters AA, Kenny PA, Roberts-Thomson SJ, Monteith GR (2013) Mitochondrial calcium uniporter silencing potentiates caspase-independent cell
PLOS ONE | www.plosone.org
death in MDA-MB-231 breast cancer cells. Biochemical and biophysical research communications. 41. Das A, Pushparaj C, Herreros J, Nager M, Vilella R, et al. (2013) T-type calcium channel blockers inhibit autophagy and promote apoptosis of malignant melanoma cells. Pigment cell & melanoma research 26: 874–885. 42. Jang SJ, Choi HW, Choi DL, Cho S, Rim H-K, et al. (2013) In vitro cytotoxicity on human ovarian cancer cells by T-type calcium channel blockers. Bioorganic & medicinal chemistry letters 23: 6656–6662. 43. Altamirano F, Valladares D, Henrı´quez-Olguı´n C, Casas M, Lo´pez JR, et al. (2013) Nifedipine treatment reduces resting calcium concentration, oxidative and apoptotic gene expression, and improves muscle function in dystrophic mdx mice. PloS one 8: e81222.
9
September 2014 | Volume 9 | Issue 9 | e105801
Abstract Purpose: Studies on the association between the use of calcium channel blockers (CCBs) and breast cancer risk have reported inconsistent results. We quantitatively assessed this association by conducting a meta-analysis based on the evidence from observational studies. Methods: We searched PubMed, MEDLINE, EMBASE and the Cochrane Library for relevant studies published up to and including December 31, 2013. We calculated pooled risk ratios (RRs) for cancer risk. Results: A total of 17 studies (9 cohort studies, 8 case-control studies) were selected for further study. These studies included 149,607 female subjects, of which 53,812 were CCBs users, who were followed for 2–16 years. The risks of breast cancer among patients receiving CCBs were significantly different for the pooled RRs (95% confidence interval) of cohort studies 1.08 (0.95, 1.20) and case-control studies 0.98 (0.86, 1.09). Differences were also noted for cancer risk, for CCBs use of ,5 years 0.96 (0.78, 1.15), and for .5 years 1.01 (0.74, 1.28), as well as for ever used 1.08 (0.95, 1.20), and for current use 1.13 (0.83, 1.42). The RR for studies longer than 10 years was 1.71 (1.01, 2.42), and for studies evaluating nifedipine was 1.10 (0.87, 1.33) and diltiazem was 0.75 (0.40, 1.10). Conclusions: The long-term use of CCBs appears to have a significant relationship with breast cancer. Well-designed clinical trials are needed to optimize the doses and types of these drugs needed to minimize their carcinogenic potential. Citation: Li W, Shi Q, Wang W, Liu J, Li Q, et al. (2014) Calcium Channel Blockers and Risk of Breast Cancer: A Meta-Analysis of 17 Observational Studies. PLoS ONE 9(9): e105801. doi:10.1371/journal.pone.0105801 Editor: Rodney John Scott, University of Newcastle, Australia Received March 4, 2014; Accepted July 25, 2014; Published September 3, 2014 Copyright: ß 2014 Li et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work is supported by Innovation Project of Shanghai Municipal Education Commission (12YZ057), the Fundamental Research Funds for the Central Universities (78210021) and the Chenguang Program from Shanghai Municipal Education Commission (10CG25). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * Email: [email protected] " WL is the first author on this work.
CCBs, which can inhibit apoptosis and thus facilitate the division of cells with malignant potential [12], have been found to increase breast cancer risk in some studies. In 1996, Pahor et al., found that in patients who had received CCBs the risk of breast cancer was increased [13]. However, the data is equivocal with some studies lending support to this association between CCBs and breast cancer risk [14–17], but not others [18–23]. Therefore the debate has been fuelled by conflicting data. We conducted a meta-analysis of observational studies of CCBs, to examine their effect on the occurrence of breast cancers.
Introduction Breast cancer is by far the most common cancer diagnosed in women (ranking second for both sexes combined) and the most common cause of death in women (ranking fifth for both sexes combined) worldwide [1,2]. The age standardized incidence increased by 40% from 1979 to 1992 in England and Wales [3]. Breast cancer is now also the leading cause of cancer-related mortality among females in economically developing countries, a shift from the previous decade during which the most common cause of cancer death was cervical cancer [4]. Hence, there is a growing interest in breast cancer prevention and several strategies have been evaluated with some promising results. Calcium channel blockers (CCBs) are prescribed primarily for treatment of hypertension and coronary heart disease [5,6]. CCBs are also used to treat esophageal diseases [7]. These drugs are potent drugs that affect various organ systems, and can cause constipation [8], increase the risk of hemorrhage [9], and impair differentiation during embryogenesis [10]. Furthermore, cases of lupus after use of diltiazem have been reported [11]. PLOS ONE | www.plosone.org
Methods Search This meta-analysis was conducted in accordance with PRISMA(Preferred Reporting Items for Systematic Reviews and MetaAnalyses) guidelines [24,25]. A comprehensive literature search was carried out using PubMed database, MEDLINE, EMBASE, China Academic Journals Full-text Database, and the Cochrane Library. We used the term ‘‘breast cancer’’ or ‘‘cancer’’ or 1
September 2014 | Volume 9 | Issue 9 | e105801
A Meta-Analysis of 17 Observational Studies
‘‘cancer(s)’’ or ‘‘neoplasm(s)’’ or ‘‘malignancy(ies)’’ or ‘‘carcinoma’’ in combination with ‘‘calcium channel blockers’’ or ‘‘verapamil’’ or ‘‘diltiazem’’ or ‘‘nifedipine’’ or ‘‘dihydropyridines’’ or ‘‘amlodipine’’ to identify the studies related to CCBs and breast cancer risk. The reference lists of all eligible articles and reviews were also scanned to identify additional relevant studies. All cancer studies were included to ensure that no results for breast cancer were overlooked that might have been reported as part of a larger study that included other cancer types.
performed according to (i) study design (cohort and case-control), (ii) duration of use, (iii) time to study observation (ever and current), and (iv) drug type to examine the impact of these factors on the association between CCB use and breast cancer. To test the robustness of this association, we performed a sensitivity analysis by excluding one study at a time. Cumulative meta-analysis was also performed to identify the change in trend of reporting risk over time.
Results
Selection/Study Characteristics
Search results
The studies considered in this meta-analysis were all observational (cohort or case-control) studies that evaluated exposure to CCBs and risk of breast cancer. We included all articles irrespective of publication length; for example, articles published as short reports or conference abstracts, even though the critical appraisal of such publications is limited, were included. Articles were excluded from the analysis if they had insufficient published data for determining an estimate of relative risk (RR) and confidence interval (CI). When there were multiple publications from the same population, only data from the most-recent report were included in the meta-analysis and remaining data were excluded. Studies reporting different measures of RR [e.g. risk ratio, rate ratio, and odds ratio (OR)] were included in the metaanalysis. In practice, these measures of effect yield a similar estimate of RR, since the absolute risk of breast cancer is low.
A total of 128 abstracts were identified and screened, and 108 studies were reviewed in detail. After excluding laboratory studies, ecological studies, and reviews, 20 studies were initially selected for analysis [13–23,29–37]. Three of these studies were excluded because they did not provide sufficient information [35–37], leaving 17 in the final analysis (Figure 1) [13–23,29–34].
Qualitative Summary The characteristics of the selected studies are shown in Table 1. Overall, 17 relevant studies (9 cohort and 8 case-control) published between 1996 and 2013 were identified. A total of 149,607 female subjects, including 53,812 CCBs users, were enrolled in these studies and followed for 2 to 16 years. Nine studies were from the United States [13,14,16,17,19–21,31,33], four from the UK [15,18,22,29], two from Denmark [23,30], and two from Canada [32,34]. Nine studies were cohort studies [13,15,18– 20,23,30,33,34], while seven were case-control studies [14,17,19,21,22,29,31], and one was a nested case-control study [32]. Of the 17 studies [13–23,29–34], 7 were population-based and 10 were hospital-based. All studies were controlled for potential confounding factors (at least for age) by matching or adjustments. Further, 7 studies [14,17–19,29,31,32] reported OR estimates for the association between CCB use and the risk of breast cancer, while 10 [13,15–17,20–23,30,34] studies reported RR. All studies had a median follow-up time of at least 3 years, while 2 studies had a median follow-up of more than 10 years.
Data abstraction We reviewed each full-text report to determine its eligibility and extracted and tabulated all the relevant data independently. The extracted data included the characteristics of the subjects (including age, number and treatment), study design, published year, follow-up period, journal of article and the covariate adjustment. Study authors were contacted as needed to obtain detailed data. Any disagreement was resolved by a consensus among the investigators.
Quantitative data synthesis The RRs and ORs with their 95% CIs were extracted from individual studies for the highest versus the lowest quartiles, and the OR from case-control studies were assumed as the estimate of the RR value in the CCBs intake meta-analysis studies. Both the fixed-effects and random-effects models were used to calculate the pooled estimate and its 95% CI. If more than one risk estimate was provided in a study that had been stratified by covariates, the estimates were pooled before data were entered into the final analysis. The heterogeneity of the data was quantified by the Q statistic and in combination with the I2 statistic, which represents the percentage of variability across studies that is attributable to heterogeneity rather than chance. Heterogeneity among studies was considered significant when P,0.05 for the Q statistic or when the I2 value was more than 50%. If there was significant heterogeneity among the studies, the random-effects model was used, otherwise, the fixed-effects model was acceptable. Publication bias was represented by funnel plots and was further assessed by the Egger test and Begg test [26]. When there was a significant publication with regard to CCB intake and breast cancer risk, the trim and fill method was applied to correct the publication bias [27,28]. All the statistical analyses were performed with STATA12.0 (StataCorp, College Station, TX, USA). We first estimated the risk of breast cancer in CCB users compared to the non-users. We used the available data from previously collected studies, which reported RR estimates for these particular associations. Pre-specified subgroup analyses were PLOS ONE | www.plosone.org
Figure 1. Flow diagram of studies included in the metaanalysis. doi:10.1371/journal.pone.0105801.g001
2
September 2014 | Volume 9 | Issue 9 | e105801
PLOS ONE | www.plosone.org
case control
cohort
cohort
Rosenberg et al. [21] USA, 1998
Michels et al. [20] US, 1998
Sørensen et al. [19] USA, 2000
3
September 2014 | Volume 9 | Issue 9 | e105801
case control
nested case-control
cohort
case control
cohort
Davis et al. [29] USA, 2007
Assimes et al. [30] Canada, 2008
Saltzman et al. [31] USA, 2013
Li et al. [13] USA, 2013
Holmes et al. [32] Canada, 2013
Cancer Epidemiology
JAMA Internal Medicine
5
.10
12
7
pharmacoepidemiology and drug safety Cancer Causes Control
3
5.7
7
3
5
3.2
6
16
5
3
5
5
Follow-up (years)
Eur J Epidemiol
BreastCancer Research and Treatment
pharmacoepidemiology and drug safety
CANCER
Arch InternMed
CANCER
CANCER
JAMA
J Hypertens
CANCER
Hypertension
THE LANCET
THE LANCET
Journal of article
63
55–74
$65
71.8
20–74
50–67
30–79
65–79
.59
66.4
60.8
40–69
51.7
72.4
NA
71.3
79.3
Age (years)
4019
2851
3201
77887
1247
49950
21177
1001
925
11726
18635
6641
2770
3198
17911
1096
3259
Number of patients
4019
1960
392
1136
600
4381
3333
512
190
11726
2361
2893
1129
759
17911
221
289
No.cases
NA
Dihydropyridines Diltiazem Verapamil Nifedipine
NA
NA
NA
Dihydropyridines Diltiazem Verapamil Nifedipine
Amlodipine Nifedipine Diltiazem
Diltiazem Verapamil Dihydropyridines
Dihydropyridines Diltiazem Verapamil Nifedipine
Dihydropyridines Diltiazem Verapamil
NA
Verapamil, Nifedipine Diltiazem
Dihydropyridines Diltiazem Verapamil
Verapamil, Nifedipine Diltiazem
Dihydropyridines Diltiazem Verapamil
Amlodipine Verapamil, Nifedipine Diltiazem
Verapamil Diltiazem Nifedipine
Study drug
AHT, antihypertensivemedications, HRT, hormone replacement therapy, BMI, body mass index, NA, not available, NSAID,nonsteroidal antiinflammatory drugs. doi:10.1371/journal.pone.0105801.t001
cohort
Fryzek et al. [28] Denmark,2006
case-control
cohort
Hole et al. [15] UK, 1998
Gonza´lez-Pe´rez et al. [27] UK, 2004
cohort
Fitzpatrick et al. [16] USA, 1997
case-control
cohort
Olsen et al. [23] Danish, 1997
Li et al. [14] USA, 2003
case control
Jick et al. [22] UK, 1997
case control
cohort
Pahor et al. [13] USA, 1996
Meier et al. [18] UK, 2000
Analysis design
References Country, year
Table 1. Characteristic of studies included in the meta-analysis.
age, stage at diagnosis history of previous cancer, and urban/rural residence
age, reference year, county, race/ ethnicity, and recency of alcohol use
age, alcohol use, income, age at menopause, waist-hip ratio; all compared against common reference group of never AHT users
sex, year of entry, age at entry, and type of antagonist
sex, year of entry, age at entry, and type of antagonist
age, calendar period, HRT use, NSAID use, parity, and age at first birth
age, calendar year, BMI, alcohol intake, smoking status, HRT use, prior breast lump and/or biopsy, hypertension
age,BMI, receipt of HRT, alcohol use, smoking status
age BMI,smoking status, alcoholism,hysterectomy, breast lumps
sex, year of entry, age at entry, and type of antagonist
history of benign breast disease, age at menarche, parity, age at first birth, and age at menopause
sex, year of entry, age at entry, and type of antagonist
sex, year of entry, age at entry, and type of antagonist
age, race, parity, age at menopause,and self-reported diabetes
sex, year of entry, age at entry, and type of antagonist
age, dose,and duration of use
age, sex, ethnic origin,heart failure, number of hospital admissions, cigarette smoking, and alcohol intake.
Covariate adjustment
A Meta-Analysis of 17 Observational Studies
A Meta-Analysis of 17 Observational Studies
a statistically contradictory effect revealed (summary RR = 1.71, 95%CI: 1.01, 2.42, for Pheterogeneity = 0.379, I2 = 0.0%). Furthermore, when subgroup analysis was performed according to different drugs, nifedipine and diltiazem the pooled RR for the studies which used nifedipine was 1.10 (95% CI: 0.87, 1.33) and studies which used diltiazem was 0.75 (95%CI: 0.40, 1.10). As shown in Figure 7, there was no evidence of heterogeneity (Q = 9.47, P = 1.000, I2 = 0.0% and Q = 4.23, P = 0.627, I2 = 0.0%, for nifedipine and diltiazem, respectively).
Quantitative Summary (Meta-analysis) We found no association between CCB use and risk of breast cancer in 12 studies (RR = 1.02, 95%CI: 0.94, 1.11) presented in Figure 2. However, there was high evidence of a lack of heterogeneity among these studies (Pheterogeneity = 0.168, I2 = 28.2%). Stratification by study design showed that the association was neutral in cohort studies (RR = 1.08, 95%CI: 0.95, 1.20) and there was a non-significant inverse association in case-control studies (RR = 0.98, 95% CI: 0.86, 1.09) [P = 0.02]. Moreover, in the subgroup analysis based on the duration of drug use, no differences were detected in duration of CCB use and breast cancer risk: ,5 years (RR = 0.96, 95%CI: 0.78, 1.15), .5 years (RR = 1.01, 95%CI: 0.74, 1.28), and ever used (RR = 1.08, 95%CI: 0.95, 1.20). However, the pooled RR for current use was 1.13 (95%CI: 0.83, 1.42) (Figure 3). The combined RR was 1.00(95%CI: 0.99,1.00) for the 3 studies which followed patients for .10 years indicating that there was no association between CCB use and breast cancer (I2 = 71.3%, Figure 4). The heterogeneity appeared to be driven by the effect of the study by Hole et al. [15] (Figure 5). When it was excluded from the analysis, the high heterogeneity was eliminated (Figure 6), and
Assessment of Publication bias The funnel plots for the use of CCBs and breast cancer did not reveal evidence of obvious asymmetrical publication bias. Furthermore, the Begg’s funnel plot (Figure 8) based on the selected studies, calculated P = 0.068 in Egger’s test and P = 0.055 egg’s test, respectively. Therefore, there is no evidence of publication bias in the analysis. However, this is indicative of some degree of publication bias, particularly in breast cancer where likely much more data exists on the topic in other studies that did not present these associations in their papers.
Figure 2. The Forest plot of relative ratios (RR) for breast cancer incidence rate between CCB users and non-users in cohort and case-control studies. Horizontal lines represent 95%CI. The diamond (and dash vertical line) represents the overall RR estimate, with the 95%CI given by its width. The solid vertical line is at the null value (RR = 1). doi:10.1371/journal.pone.0105801.g002
PLOS ONE | www.plosone.org
4
September 2014 | Volume 9 | Issue 9 | e105801
A Meta-Analysis of 17 Observational Studies
Figure 3. Meta-analysis and pooled relative ratios (RR) of the time-related CCBs use and breast cancer occurrence. doi:10.1371/journal.pone.0105801.g003
breast cancer. These findings suggest that longer follow-up maybe required to show any association between CCB use and breast cancer. In our subgroup analyses, the results were not substantially affected by study design. Cohort and case-control studies alone showed no association between CCBs use and risk of breast cancer. Nevertheless, there was deviation in the subgroup analyses which evaluated the different CCBs (nifedipine and diltiazem). The test of interaction was not statistically significant in subgroup analyses but was significant among subgroups representing nifedipine use. Although, biological mechanisms through which calcium-channel blockers could influence breast cancer risk are unknown, CCBs may inhibit apoptosis through increasing intracellular calcium. The results that the increased risk with nifedipine, and that an decreased risk associated with diltiazem may help inform studies aimed at elucidating potential biological mechanisms. In short, we found an association between CCB use and breast cancer risk. Moreover, it has been suggested that CCBs increase
Discussion Over the past few decades, there has been increasing understanding of the role of CCBs in the development of breast cancer. The present meta-analysis of 17 observational studies, indicates that there is no increase in breast cancer risk among CCBs users as compared to non-users. This association remained stable even after the sensitivity analysis. Overall, when compared to non-users of CCBs, we found no significant difference in breast cancer risk among ever users, use for ,5 years and use for .5 years. However, an inverse association was observed that suggested there is a 13% increase in the risk of breast cancer occurrence in current CCB users compared to nonusers. There was significant heterogeneity among studies with a duration of longer than 10 years. Nevertheless, the heterogeneity was eliminated when Hole et al. was removed from the analysis [15]. Furthermore, an inverse association was documented and the combined RR was 1.71%. This result provides evidence that longterm use of CCBs may be associated with an increased risk of
PLOS ONE | www.plosone.org
5
September 2014 | Volume 9 | Issue 9 | e105801
A Meta-Analysis of 17 Observational Studies
Figure 4. The combined relative ratios (RR) and 95%CI for the 3 studies which had a duration of longer than 10 years. doi:10.1371/journal.pone.0105801.g004
studies, it is uncertain whether the cases are comparably representative. Moreover, both the funnel plot and Egger’s test do not have high enough power to detect the bias. Also, no Asian was included in our analysis. Finally, the use of CCBs differed across the studies, and some of these studies did not assess or adjust for enough potential confounding variables. However, potential publication bias could be of concern because small studies with null results tended not to be published, especially in the case of clinical trials. In our meta-analysis, we found no evidence of
the risk of cancer by inhibiting apoptosis, or programmed cell death, by DNA fragmentation of dysfunctional and old cells [38]. The CCBs might affect cancer risk generally or be limited to specific sites where calcium mechanisms predominate. Intracellular calcium ion balance regulates apoptosis in vitro [39]; for example, mitochondrial calcium uniporter silencing potentiates caspase-independent cell death in MDA-MB-231 breast cancer cells [40]. In addition, calcium antagonists inhibit apoptosis in vitro [41,42]. Animal data suggest that nifedipine reduced resting calcium concentration and apoptotic gene expression in mice [43]. That is to say, the available laboratory evidence lends support to our findings, which indicate that the use of CCBs increases breast cancer risk. What’s more, it is feasible that breast tissue may be more vulnerable to alterations in apoptotic activity than the other types of tissue. In complex secretory tissue such as the mammary gland, a complex relationship between apoptosis and breast carcinoma exists that may hormonally related. Therefore, more related studies are needed to illustrate this hypothesis.
Strengths and limitations The strength of the present analysis lies in inclusion of 17 observational studies, reporting data from more than 149,607 female participants from multiple nations and was performed with a high level of precision, including 53,812 CCBs user. In addition, this was the first study to use a meta-analysis to investigate the use of CCBs and breast cancer risk. In the analysis of cancer incidence, there was no significant difference in RR between the case-control studies and the cohort studies. Our meta-analysis has several limitations. First, we did not search for unpublished studies or for original data. Second, the included studies were different although we did not detect significant publication bias between PLOS ONE | www.plosone.org
Figure 5. The sensitivity analysis for the 3 studies followed-up for more than ten years. doi:10.1371/journal.pone.0105801.g005
6
September 2014 | Volume 9 | Issue 9 | e105801
A Meta-Analysis of 17 Observational Studies
Figure 6. The combined relative ratios (RR) and 95%CI for the 2 studies followed-up more than ten years. doi:10.1371/journal.pone.0105801.g006
Figure 7. The combined relative ratios (RR) and 95%CI for the 3 studies which state the use of nifedipine or diltiazem. doi:10.1371/journal.pone.0105801.g007
PLOS ONE | www.plosone.org
7
September 2014 | Volume 9 | Issue 9 | e105801
A Meta-Analysis of 17 Observational Studies
Conclusions In conclusion, the long-term use of CCBs appears to have a significant relationship with breast cancer. These findings provide support for the appropriate use of CCBs for those patients who have potentially increased risk of breast cancer. However, more well-designed clinical trials are needed to determine the effect of CCBs on breast cancer, and to optimize the doses and types of these drugs needed to minimize their carcinogenic potential.
Supporting Information Checklist S1 PRISMA Checklist.
(PDF)
Author Contributions
Figure 8. Begg’s funnel plot of studies included in the analysis. The funnel plot displays log odds ratio (OR) against its standard error (s.e.) for each individual study. The horizontal line represents the estimate of the overall OR, with the dash lines indicate the expected 95% CI for a given s.e. doi:10.1371/journal.pone.0105801.g008
Conceived and designed the experiments: WL QS. Performed the experiments: WL JL. Analyzed the data: WL WW. Contributed reagents/materials/analysis tools: WL WW. Wrote the paper: WL QS WW JL QL FH.
publication bias. Therefore, we will update our study when possible.
References 18. Meier CR, Derby LE, Jick SS, Jick H (2000) Angiotensin-converting enzyme inhibitors, calcium channel blockers, and breast cancer. Archives of internal medicine 160: 349. 19. Sørensen HT, Olsen JH, Mellemkjær L, Thulstrup AM, Steffensen FH, et al. (2000) Cancer risk and mortality in users of calcium channel blockers. Cancer 89: 165–170. 20. Michels KB, Rosner BA, Walker AM, Stampfer MJ, Manson JE, et al. (1998) Calcium channel blockers, cancer incidence, and cancer mortality in a cohort of US women. Cancer 83: 2003–2007. 21. Rosenberg L, Rao RS, Palmer JR, Strom BL, Stolley PD, et al. (1998) Calcium channel blockers and the risk of cancer. JAMA: The Journal of the American Medical Association 279: 1000–1004. 22. Jick H, Jick S, Derby LE, Vasilakis C, Myers MW, et al. (1997) Calcium-channel blockers and risk of cancer. The Lancet 349: 525–528. 23. Olsen JH, Sørensen HT, Friis S, McLaughlin JK, Steffensen FH, et al. (1997) Cancer risk in users of calcium channel blockers. Hypertension 29: 1091–1094. 24. Knobloch K, Yoon U, Vogt PM (2011) Preferred reporting items for systematic reviews and meta-analyses (PRISMA) statement and publication bias. Journal of Cranio-Maxillofacial Surgery 39: 91–92. 25. Moher D, Liberati A, Tetzlaff J, Altman DG (2009) Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Annals of internal medicine 151: 264–269. 26. Egger M, Smith GD, Schneider M, Minder C (1997) Bias in meta-analysis detected by a simple, graphical test. Bmj 315: 629–634. 27. Peters JL, Sutton AJ, Jones DR, Abrams KR, Rushton L (2007) Performance of the trim and fill method in the presence of publication bias and between-study heterogeneity. Statistics in medicine 26: 4544–4562. 28. Duval S, Tweedie R (2000) Trim and Fill: A Simple Funnel-Plot–Based Method of Testing and Adjusting for Publication Bias in Meta-Analysis. Biometrics 56: 455–463. 29. Gonzalez-Perez A, Ronquist G, Garcia Rodriguez LA (2004) Breast cancer incidence and use of antihypertensive medication in women. Pharmacoepidemiology and drug safety 13: 581–585. 30. Fryzek JP, Poulsen AH, Lipworth L, Pedersen L, Nørgaard M, et al. (2006) A cohort study of antihypertensive medication use and breast cancer among Danish women. Breast cancer research and treatment 97: 231–236. 31. Davis S, Mirick DK (2007) Medication use and the risk of breast cancer. European journal of epidemiology 22: 319–325. 32. Assimes TL, Elstein E, Langleben A, Suissa S (2008) Long-term use of antihypertensive drugs and risk of cancer. Pharmacoepidemiology and drug safety 17: 1039–1049. 33. Saltzman BS, Weiss NS, Sieh W, Fitzpatrick AL, McTiernan A, et al. (2013) Use of antihypertensive medications and breast cancer risk. Cancer Causes & Control 24: 365–371. 34. Holmes S, Griffith EJ, Musto G, Minuk GY (2013) Antihypertensive medications and survival in patients with cancer: A population-based retrospective cohort study. Cancer epidemiology 37: 881–885. 35. Beiderbeck-Noll A, Sturkenboom M, Van Der Linden P, Herings R, Hofman A, et al. (2003) Verapamil is associated with an increased risk of cancer in the elderly: the Rotterdam study. European Journal of Cancer 39: 98–105.
1. Siegel R, Naishadham D, Jemal A (2012) Cancer statistics, 2012. CA: a cancer journal for clinicians 62: 10–29. 2. Hunter DJ, Reddy KS (2013) Noncommunicable diseases. New England Journal of Medicine 369: 1336–1343. 3. Quinn M, Allen E (1995) Changes in incidence of and mortality from breast cancer in England and Wales since introduction of screening. Bmj 311: 1391– 1395. 4. Jemal A, Bray F, Center MM, Ferlay J, Ward E, et al. (2011) Global cancer statistics. CA: a cancer journal for clinicians 61: 69–90. 5. Chobanian AV (1996) Calcium channel blockers. JAMA: the journal of the American Medical Association 276: 829–830. 6. Psaty BM, Heckbert SR, Koepsell TD, Siscovick DS, Raghunathan TE, et al. (1995) The risk of myocardial infarction associated with antihypertensive drug therapies. JAMA: the journal of the American Medical Association 274: 620– 625. 7. Konrad-Dalhoff I, Baunack A, Ra¨msch K, Ahr G, Kraft H, et al. (1991) Effect of the calcium antagonists nifedipine, nitrendipine, nimodipine and nisoldipine on oesophageal motility in man. European journal of clinical pharmacology 41: 313–316. 8. Pahor M, Guralnik J, Chrischilles EA, Wallace RB (1994) Use of laxative medication in older persons and associations with low serum albumin. Journal of the American Geriatrics Society 42: 50–56. 9. Wagenknecht L, Furberg C, Hammon J, Legault C, Troost B (1995) Surgical bleeding: unexpected effect of a calcium antagonist. BMJ: British Medical Journal 310: 776. 10. Pahor M, Carbonin P, Guralnik J, Havlik R, Furberg C (1996) Risk of gastrointestinal haemorrhage with calcium antagonists in hypertensive persons over 67 years old. The Lancet 347: 1061–1065. 11. Stein G, Srivastava MK, Merker H-J, Neubert D (1990) Effects of calcium channel blockers on the development of early rat postimplantation embryos in culture. Archives of toxicology 64: 623–638. 12. Carson DA, Ribeiro JM (1993) Apoptosis and disease. The Lancet 341: 1251– 1254. 13. Pahor M, Guralnik JM, Ferrucci L, Corti M-C, Salive ME, et al. (1996) Calcium-channel blockade and incidence of cancer in aged populations. The Lancet 348: 493–497. 14. Li CI, Malone KE, Weiss NS, Boudreau DM, Cushing-Haugen KL, et al. (2003) Relation between use of antihypertensive medications and risk of breast carcinoma among women ages 65–79 years. Cancer 98: 1504–1513. 15. Hole DJ, Gillis CR, McCallum IR, McInnes GT, MacKinnon PL, et al. (1998) Cancer risk of hypertensive patients taking calcium antagonists. Journal of hypertension 16: 119–124. 16. Fitzpatrick AL, Daling JR, Furberg CD, Kronmal RA, Weissfeld JL (1997) Use of calcium channel blockers and breast carcinoma risk in postmenopausal women. Cancer 80: 1438–1447. 17. Li CI, Daling JR, Tang M-TC, Haugen KL, Porter PL, et al. (2013) Use of antihypertensive medications and breast cancer risk among women aged 55 to 74 years. JAMA internal medicine 173: 1629–1637.
PLOS ONE | www.plosone.org
8
September 2014 | Volume 9 | Issue 9 | e105801
A Meta-Analysis of 17 Observational Studies
36. Braun S, Boyko V, Behar S, Reicher-Reiss H, Laniado S, et al. (1998) Calcium Channel Blocking Agents and Risk of Cancer in Patients With Coronary Heart Disease 1. Journal of the American College of Cardiology 31: 804–808. 37. Lindberg G, Lindblad U, Lo¨w-Larsen B, Merlo J, Melander A, et al. (2002) Use of calcium channel blockers as antihypertensives in relation to mortality and cancer incidence: a population-based observational study. pharmacoepidemiology and drug safety 11: 493–497. 38. Moon D, McCormack D, McDonald D, McFadden D (2013) Pterostilbene induces mitochondrially derived apoptosis in breast cancer cells in vitro. Journal of Surgical Research 180: 208–215. 39. Handayani S (2013) Selaginella Active Fractions Induce Apoptosis on T47D Breast Cancer Cell. Indonesian Journal of Pharmacy 23. 40. Curry MC, Peters AA, Kenny PA, Roberts-Thomson SJ, Monteith GR (2013) Mitochondrial calcium uniporter silencing potentiates caspase-independent cell
PLOS ONE | www.plosone.org
death in MDA-MB-231 breast cancer cells. Biochemical and biophysical research communications. 41. Das A, Pushparaj C, Herreros J, Nager M, Vilella R, et al. (2013) T-type calcium channel blockers inhibit autophagy and promote apoptosis of malignant melanoma cells. Pigment cell & melanoma research 26: 874–885. 42. Jang SJ, Choi HW, Choi DL, Cho S, Rim H-K, et al. (2013) In vitro cytotoxicity on human ovarian cancer cells by T-type calcium channel blockers. Bioorganic & medicinal chemistry letters 23: 6656–6662. 43. Altamirano F, Valladares D, Henrı´quez-Olguı´n C, Casas M, Lo´pez JR, et al. (2013) Nifedipine treatment reduces resting calcium concentration, oxidative and apoptotic gene expression, and improves muscle function in dystrophic mdx mice. PloS one 8: e81222.
9
September 2014 | Volume 9 | Issue 9 | e105801
